#0: Fix double deallocate

tenstorrent · Nov 12, 2024 · b16ff97 · b16ff97
1 parent 274f58a
commit b16ff97
Showing 1 changed file with 0 additions and 1 deletion.
diff --git a/models/demos/llama3/tt/llama_attention.py b/models/demos/llama3/tt/llama_attention.py
@@ -351,7 +351,6 @@ def forward_decode(
             dense_out_sharded, ttnn.L1_MEMORY_CONFIG
         )  # TODO: remove as soon as we have sharded support in for all CCL
 
-        ttnn.deallocate(attn_output_cat)
         ttnn.deallocate(dense_out_sharded)
 
         # All reduce